Full text loading...
- ISSN: 2202-0586
- E-ISSN:
-
oa A Bayesian inference tool for geophysical joint inversions
- Australian Society of Exploration Geophysicists
- Source: ASEG Extended Abstracts, Volume 2016, Issue 25th International Conference and Exhibition – Interpreting the Past, Discovering the Future, Dec 2016, p. 1 - 10
-
- 01 Dec 2016
Abstract
Geophysical joint inversions seek to exploit the statistical fact that a model that simultaneously satisfies two or more independent data sets is more likely to represent geological ‘reality’ than a model that only satisfies a single data set. Interpreting geophysical data directly rapidly exceeds the capacity of a human as more data are added, so some form of machine assistance is usually required. Conventional inversion techniques can produce a ‘best fit’ model but this might only be one of a large range of possible models that fit the data. Bayesian inference provides a tool to evaluate the relative probability of all possible geological models in a given set, thereby quantifying the amount of information the data is actually providing.
Over 2012-2014, National ICT Australia (NICTA; now Data 61) worked with a number of university, government and industry partners, with support from the Australian Renewable Energy Agency, to build a Bayesian inference software tool for geophysical joint inversions. The tool was initially directed at geothermal energy exploration but is equally applicable to investigating other geological problems. For one geothermal exploration problem, Bayesian inference allowed us to jointly invert gravity, magnetics, magnetotelluric soundings and borehole temperature records to map in three dimensions the probability of encountering granite >270°C beneath the Moomba region of South Australia. The results correlated well with an independent deterministic inversion carried out by Geoscience Australia, but provided a much richer interpretation in probability space.
NICTA released the software tools as open source code on the GitHib platform.
© 2016 ASEG